• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Subscribers 47 subscribers
  • Views 911 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Powering the DDC232

John Ladasky
Prodigy 225 points
Other Parts Discussed in Thread: DDC232

I am developing a device which will need to measure multiple photodiodes producing current in the 400 fA range, with as much precision as possible.  I am investigating the suitability of the DDC232 for this task.

My signals are essentially DC levels, so I can integrate signals for as long as the analog front end IC will allow.  The documentation for the DDC232 says that I can integrate for a full second, and that is what I plan to do.

That being said, my device will be battery-powered, and thus it will have very limited power supply choices.  If I use a 3.6V lithium battery, I will need to use a PMIC with a boost rail to obtain the 5.0V analog supply for the DDC232, and also to power the 4.096V reference.  Is this advisable?  The power supply rejection ratio specification for the DDC232 appears to be 100-800 ppm at DC (-40 to -31 dB).  Those PSRR numbers don't seem very good to me, but I don't know whether I am interpreting them correctly.  If I want to preserve the DDC232's 20 bits (120 dB) of resolution, do I really need to get my power supply noise 80 - 89 dB below my 400 fA signal?  That seems impossible to me, even for an LDO.  (Of course I will also shield the photodiodes and the DDC232 itself, and pay careful attention to my board layout to minimize noise.)

Thanks for any advice you may have.

  • 0
    TI__Mastermind 49756 points
    Hello John,

    We have received your inquiry about using the DDC232 for your application and have assigned the apps expert to the post.
    He will get back to you as soon as possible.
  • 0 in reply to Praveen Aroul
    Prodigy 225 points

    Thank you, I look forward to the reply.

  • 0 in reply to John Ladasky
    Prodigy 225 points
    Dear Mr. Aroul.

    I was hoping to provide an update to my supervisor at our weekly meeting regarding our search for an analog front end. Please let me know the status of our inquiry. Thanks!
  • 0 in reply to John Ladasky
    TI__Expert 4680 points

    Hi John,

    Sorry for late reply... The way to interpret the PSRR is "offset PSRR", i.e., how much the offset would change on the device (100-800ppm on this case) for every V changed on the supply. Notice that this is for DC-like changes on the supply. Also notice that this was characterized for the conditions on top of the table (like Range 7, 350pC, and fast integration times). 

    On that sense, a quick check shows that the sensibility of the device may be good enough with any range. With 400fA and integrating the maximum specified on the device (1s) you would be getting 400fC. I don't know what is your input capacitance, but noise is ~1fCrms on the most sensitive range (Range 0) and although it degrades on Range 7, you may still be ok, depends on the target SNR. Also noise does increase with integration time, but probably still ok (we do not have a graph for that on the 232 DS but you can see one on the 264 DS to get a sense for it...)

    So, say, for the sake of this exercise that we stay with the Range 7, the noise would be about 5-20ppm of FSR. A supply variation of 100mV would yield about 10-80ppm change. I am not an expert on supply design but it looks to me like you got some good room (?). I do not know how the PSRR will change if you move to a more sensitive range, but may be that the absolute variation is similar, so, effectively increasing in ppm. It may still be worth doing so, but I guess needs to be checked...

    A more concerning thing is the Ibias listed as 100fA typical but can go up to 10000fA. Again, this is for Range 7 but probably will not change significantly at Range 0 (I do not know but a guess). Also, this Ibias changes with temperature. I do not know here exactly but a factor 2 for every 10C is reasonable. So, it is pretty clear that you will need to calibrate it out. I.e., measure without input, measure with input, and then subtract both.

    Overall, I recall a user that in a lab instrument was measuring fAs with this device, using the most sensitive range and integrating for 1s, but he really worked every non-ideality in the system... 

    Best regards,

    Eduardo

  • 0 in reply to Eduardo Bartolome
    Prodigy 225 points
    Dear Mr. Bartolome,

    Thank you for your reply. It sounds like my understanding of the PSRR measurement was more or less accurate. However, you have good suggestions about how to measure bias current and other noise. My supervisor and I have decided to give it a try. We are going to power the device with LDO rails after all in order to avoid problems with power supply ripple.